{"title":"Research and Optimization of Urban Rail Transit Handover Based on WLAN","authors":"Wang Chenglong, Wang Yeli, Wang Xinji","doi":"10.1109/ICTLE53360.2021.9525657","DOIUrl":null,"url":null,"abstract":"In order to adapt to the development trend of urban rail transit systems, the traditional IEEE 802.11 standard handover method is aimed at the problems of high handover delay and ping-pong handover. The train control system CBTC (Communication Based Train Control) handover is studied. This paper studies the WLAN (Wireless Local Area Network) handover strategy combined with DQN (Deep Q-Network) under the WLAN urban rail transit model. This strategy extracts and inputs train driving characteristic status information, and performs switching actions based on the signal strength of the serving AP and the target AP. Finally, after deep neural network training, the best switching points at different speeds are obtained. The optimal value obtained by the DQN algorithm is substituted into the model for simulation verification. The results show that compared with the traditional method, the throughput of this strategy is increased by 36%, and the packet delay is reduced by 55%. Prove the effectiveness of DQN algorithm in WLAN urban rail transit handover. At the same time, this strategy can obtain the optimal switching position after training at different speeds and different AP distances. This makes the strategy widely applicable and makes the WLAN urban rail transit switching system safer and more reliable.","PeriodicalId":199084,"journal":{"name":"2021 9th International Conference on Traffic and Logistic Engineering (ICTLE)","volume":"391 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 9th International Conference on Traffic and Logistic Engineering (ICTLE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICTLE53360.2021.9525657","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In order to adapt to the development trend of urban rail transit systems, the traditional IEEE 802.11 standard handover method is aimed at the problems of high handover delay and ping-pong handover. The train control system CBTC (Communication Based Train Control) handover is studied. This paper studies the WLAN (Wireless Local Area Network) handover strategy combined with DQN (Deep Q-Network) under the WLAN urban rail transit model. This strategy extracts and inputs train driving characteristic status information, and performs switching actions based on the signal strength of the serving AP and the target AP. Finally, after deep neural network training, the best switching points at different speeds are obtained. The optimal value obtained by the DQN algorithm is substituted into the model for simulation verification. The results show that compared with the traditional method, the throughput of this strategy is increased by 36%, and the packet delay is reduced by 55%. Prove the effectiveness of DQN algorithm in WLAN urban rail transit handover. At the same time, this strategy can obtain the optimal switching position after training at different speeds and different AP distances. This makes the strategy widely applicable and makes the WLAN urban rail transit switching system safer and more reliable.